[Ontology-editors] forward and reverse transcription

[Midori Harris]

response to latest comment; other stuff snipped for clarity

On Tue, 2 Dec 2008, Karen Christie wrote:

> one further comment inline
>
>>> 1. Based on the current defs of "RNA-dependent DNA replication" (which 
>>> includes the phrase 'RNA-dependent DNA polymerases (e.g. reverse 
>>> transcriptase)' and of "transcription, RNA-dependent" which has 'reverse 
>>> transcriptase' as a synonym, these two terms do seem to be identical.
>>> 
>>> However, it seems to me that the reverse transcription (txn) step is only 
>>> half of "RNA-dependent DNA replication" because reverse txn only makes a 
>>> ssDNA copy. This ssDNA then serves as the template for production of dsDNA 
>>> (going off my old biology book). So, perhaps the def of "RNA-dependent DNA 
>>> replication" needs to be broader to encompass the whole process, and 
>>> "transcription, RNA-dependent" could have a part_of relationship to such a 
>>> redefined/new term.
>> 
>> That's an interesting point, and one that hadn't occurred to me, since I 
>> was going by just the existing GO defs. Your proposed changes sound like 
>> they would work well, assuming that reverse transcription always happens in 
>> the context of dsDNA synthesis (I don't know of any non-experimental 
>> exceptions). Perhaps the broader process should be named RNA-dependent DNA 
>> synthesis or something to that effect, rather than use 'DNA replication' -- 
>> DNA replication seems a bit odd for a process that doesn't start with DNA.
>
> Julie and I both sort of thought that reverse txn occurs both in the contects 
> of DNA replication and mRNA production. Neither of us is a viral expert, so 
> we could be wrong. But if our current sense is right, then we probably can't 
> have the only 'reverse transcription' term be part_of 'RNA-dependent DNA 
> replication'. We might need to have a general term and child terms that can 
> receive appropriate parentage under DNA replication etc. So, it seems that we 
> need to get a better sense of the biological contexts of reverse 
> transcription before proceeding.

Fair enough. To deal with the current SF item I can move transcription out 
from under gene expression, and put DNA-dependent txn in; that will err on 
the side of caution to avoid one path that might be false.

As for where 'reverse transcription' should go, another option is to bide 
our time until we've deployed the has_part relationship type (which I hope 
won't be too long now ...), and then say that RNA-dependent RNA 
replication has_part reverse transcription, and likewise viral mRNA 
production.

m